Hi Laurent,
Laurent Pinchart wrote:
Hi Sakari,
On Tuesday 03 June 2014 15:32:24 Sakari Ailus wrote:
On Mon, Jun 02, 2014 at 05:10:02PM +0200, Laurent Pinchart wrote:
Expose the pad-level get caps, query, get and set DV timings ioctls.
Signed-off-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
---
utils/media-ctl/libv4l2subdev.c | 72 ++++++++++++++++++++++++++++++++++++
utils/media-ctl/v4l2subdev.h | 53 ++++++++++++++++++++++++++++++
2 files changed, 125 insertions(+)
diff --git a/utils/media-ctl/libv4l2subdev.c
b/utils/media-ctl/libv4l2subdev.c index 14daffa..8015330 100644
--- a/utils/media-ctl/libv4l2subdev.c
+++ b/utils/media-ctl/libv4l2subdev.c
@@ -189,6 +189,78 @@ int v4l2_subdev_set_selection(struct media_entity
*entity,
return 0;
}
+int v4l2_subdev_get_dv_timings_caps(struct media_entity *entity,
+ struct v4l2_dv_timings_cap *caps)
+{
+ unsigned int pad = caps->pad;
+ int ret;
+
+ ret = v4l2_subdev_open(entity);
+ if (ret < 0)
+ return ret;
In every function v4l2_subdev_open() is called before the ioctl command. How
about implementing a wrapper which does both, and using that before this
patch?
I'm not too fond of the current subdev API. Subdevs are opened implicitly but
must be closed explicitly. Implicit open is of course easy, but the unbalanced
API makes me feel uneasy. I'm open to ideas for better alternatives :-)
I had a patch to reference count v4l2_subdev_open() (and close()), but
it didn't get applied back in 2012. And... apparently the reason for
that was that it didn't, well, apply. :-)
<URL:http://www.spinics.net/lists/linux-media/msg52258.html>
I'll resend it.
--
Sakari Ailus
sakari.ailus@xxxxxx
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
